Sepiolite (Si12Mg8O30(OH)4(H2O)4⋅8H2O), a natural sheet silicate with large specific surface area and abundant pores (Inagaki et al., Reference Inagaki, Fukushima and Doi1990), has been used widely as an adsorbent for organic compounds such as phenanthrene (Gonzalez-Santamaria et al., Reference Gonzalez-Santamaria, Lopez and Ruiz2017) and volatile organic compounds (VOCs; Kai et al., Reference Kai, WeiMin and YunHua2011; Shen, Reference Shen2019), and as a humidity controller to regulate indoor humidity (Gonzalez et al., Reference Gonzalez, Molina-Sabio and Rodriguez-Reinoso2001; Yamaguchi, Reference Yamaguchi2012; Cho et al., Reference Cho and You2020). Due to its excellent adsorption properties, sepiolite was used as the filter tips of cigarettes in the mid-eighteenth century (Zhang, Reference Zhang2000) and more recently has been used in the famous Turkish tobacco pipes, where it can be used to reduce tar and nicotine levels. Sepiolite powders have also been used as coatings and wallpapers for regulating humidity and thus improving indoor air quality since the 1970s (Wu et al., Reference Wu, Gong and Yu2014). However, the natural sepiolite in China, referred to as sepiolite clay, is rich in impurities and is utilized only in the form of powders with low mechanical strength. Therefore, natural sepiolite cannot be used directly as a building material and its further application is greatly limited. The best way to tackle this problem is to convert the sepiolite clay into a cemented sepiolite aggregate for further use.
To obtain mechanically strong building materials, the traditional sintering method (≥1173.15 K) is used widely. The inherent properties and performance of the original materials might be preserved when the processing temperature does not exceed 773.15 K (Ishida, Reference Ishida2002). Therefore, transformation of natural sepiolite into a tough matrix that still retains the inherent microstructure and performance of the original sepiolite is a critical step.
By mimicking the lithification processes and using the hardening mechanisms of cement and concrete, a cemented sepiolite aggregate was obtained from natural sepiolite clay under hydrothermal treatment with calcium aluminium silicate hydrate (C-A-S-H) or Al-tobermorite formation, and because of the low temperature (≤473.15 K) of hydrothermal solidification, the microstructure (porosity) of raw sepiolite might be preserved in the matrix of the sepiolite specimens (Sarıkaya et al., Reference Sarıkaya, Önal and Pekdemir2020). The hydrothermal technology has been applied widely not only in inorganic material synthesis, but also in organic material compositions (Lyu et al., Reference Lyu, Zeng, Yun, Wei and Jin2014; He et al., Reference He, Hu, Zhong, Jin, Fan, Hu and Jing2019).
Suitable relative humidity (RH) not only creates a comfortable environment, but may also restrict bacteria and virus reproduction. A RH ranging from 40 to 70% is considered to be the most comfortable for human beings (Arundel et al., Reference Arundel, Sterling and Biggin1986). The VOCs mainly emanating from indoor materials such as coatings, adhesive agents, etc. (Shang et al., Reference Shang, Du and Xu2002), may cause serious harm to human health, even at very low concentrations (Xu et al., Reference Xu, Zheng and Zhang2018), and therefore the removal of indoor VOCs is extremely important.
The adsorption behaviour of moisture and VOCs may be described using various isotherm models. According to the classification of the International Union of Pure and Applied Chemistry (IUPAC), the moisture adsorption isotherm belongs to type IV, which indicates the presence of mesopores exhibiting adsorption hysteresis characteristics. Our previous work showed that a pore size from 4.2 to 7.1 nm, such as those found in mesoporous materials, leads to effective moisture adsorption/desorption performance for humidity regulation (Lan et al., Reference Lan, Jing and Li2017). On the other hand, the isotherm of formaldehyde adsorption is of type I (Song et al., Reference Song, Qiao and Yoon2007), which exhibits mainly microporous adsorption characteristics. The diameter of formaldehyde molecules is ~0.45–0.75 nm, and based on the Horvath–Kawazoe (HK) method (Chen et al., Reference Chen and Yang1996), the effective pore size for adsorption should be as small as one or two times that of the adsorbate. Therefore, materials with micropores should have a good VOC (formaldehyde) removal efficiency (Chen et al., Reference Chen and Yang1996). As sepiolite is rich in micropores and mesopores (Sun et al., Reference Sun, Delacaillerie and Fripiat1995), aggregates of this mineral with both types of porosity would be useful as functional materials that have humidity-regulating and VOC-adsorbing capacities.
To the best of our knowledge, the hydrothermal solidification of sepiolite clay into a cemented sepiolite aggregate for humidity regulation and VOC (formaldehyde) adsorption has not been reported before. The present work aims to: (1) transform the sepiolite into a tough sepiolite-based material (cemented sepiolite aggregate) and determine its hardening mechanism; (2) study the humidity-regulating property of the cemented sepiolite aggregate and improve this humidity-regulating performance; and (3) study the VOC (formaldehyde) adsorption and its mechanism. The results are expected to provide practical information on a cemented sepiolite aggregate that might be used directly indoors for both humidity regulation and VOC adsorption.
Experimental
Sepiolite raw material
The sepiolite used in this study was obtained from HongYan Sepiolite Corporation, Ltd, in Hunan Province, China. Reagent-grade calcium hydroxide (Ca(OH)2), obtained from Sinopharm Chemical Reagent (Shanghai, China), was used as an additive. The sepiolite was ground in a ball mill to pass through a 100-mesh sieve. The particle-size distribution curve of the sepiolite is shown in Fig. 1. The main chemical components of the sepiolite (Table 1) are SiO2 (58.7 mass%) and Al2O3 (11.9 mass%). CaO (1.73 mass%) is a minor component. The semi-quantitative analysis based on the X-ray diffraction (XRD) traces (Fig. 2) shows that the main phases in the original material are sepiolite (73.1%), quartz (13.7%) and talc (10.8%). The Fourier-transform infrared (FTIR) spectrum of the sepiolite is shown in Fig. 2. The bands at 3623, 3557 and 3425 cm–1 correspond to Al–OH stretching vibrations (Ahlrichs et al., Reference Ahlrichs, Serna and Serratosa1975) and the bands centred at 1664 and 691 cm–1 are due to stretching vibrations of Mg–OH groups. The stretching bands at 1031 and 471 cm–1 belong to Si–O–Si groups of the tetrahedral sheet, while the stretching bands at 798 and 780 cm–1 belong to Si–O bonds (Fig. 2) (Liang et al., Reference Liang, Xu, Sun, Wang, Sun, Sun and Qin2011; Ozcan et al., Reference Ozcan and Gok2012).
Fig. 1. Particle-size distribution of raw sepiolite.
Fig. 2. XRD trace (left) and FTIR spectrum (right) of raw sepiolite.
Table 1. Chemical composition of raw sepiolite (mass%).
LOI = loss on ignition.
Hydrothermal solidification
The dried sepiolite powder (353.15 K, 24 h) mixed at various ratios (0–50 mass%) with Ca(OH)2 were used as starting materials. The starting materials were mixed manually with 30 mass% distilled water using a pestle and mortar, and then the mixtures were compacted by uniaxial pressure of 30 MPa in a rectangular mould (40 mm × 15 mm × 5 mm). The green specimens were treated in a Teflon-lined stainless-steel autoclave under saturated steam pressure at 373.15–513.15 K for up to 12 h, and after hydrothermal processing the specimens were dried in air at 353.15 K for 24 h before testing.
Characterization
The chemical analysis (major elements) of the raw material was performed using an X-ray fluorescence (XRF) spectrometer (Rigaku RIX 3100, Japan). The particle size of the raw material was investigated via laser diffraction analysis (LPSA, Malvern 2000) using distilled water to prepare the suspension. The hydrothermally synthesized specimens (40 mm × 15 mm × 5 mm) were used to measure the three-point flexural strength, employing the Japanese industrial standard (JIS R1601) at a loading rate of 0.5 mm min–1 using a universal testing machine (XO-106A, Xie Qiang Instrument Technology). All flexural strength values reported here are averages of three measurements obtained for three different samples.
The crushed specimens were then investigated using several analytical techniques. The crystalline phases were determined via XRD (Bruker D8 Advance Powder Diffractometer) at 40 kV and 60 mA using Cu-Kα radiation (λ = 1.54 Å, scanning range 5–60°2θ, scanning speed 5° min–1). Jade 6 software with reference data (PDF 2004 database) was utilized to identify the mineral phases and to perform the semi-quantitative analysis. The FTIR spectra of the materials were obtained by FTIR spectrometry (EQUINOXSS FTIR spectrometer) using the KBr pellet technique. The fracture surface morphology was examined using an environmental scanning electron microscope (ESEM; Quanta 200 FEG ESEM) on samples that had previously been gold-sputtered (10 nm thick). The specific surface area and total pore volume were determined using nitrogen gas sorption analysis at 77.15 K (Autosorb-1 Quantachrome Instruments, Boynton Beach, FL, USA) based on the Brunauer–Emmett–Teller (BET) equation.
The humidity-regulating apparatus (Fig. 3) and measurements were developed according to Japanese industrial standards (JIS A 1470-1, 2:2008). The surfaces of the specimens, except for their top surface, were covered with aluminium tape to ensure that the adsorbed/desorbed water vapour would only migrate through the top surface. The specimens were placed in an airtight box with a RH of 75% for 24 h, and then at a RH of 33% for another 24 h. The RHs of 75% and 33% in the airtight box were controlled by MgCl2⋅6H2O and NaCl saturated salt solutions, respectively. During this process, the mass change of the specimens due to moisture adsorption/desorption was measured every 10 min.
Fig. 3. Apparatus for measuring the humidity-regulating (left) and formaldehyde-adsorption (right) properties of the materials.
The apparatus for investigating the formaldehyde adsorption was built based on Chinese industry standards (GB/T35243-2017). The specimens were placed in the sealed apparatus, then a constant formaldehyde concentration of 1 mg m–3 was set up in the sealed apparatus, and the formaldehyde concentration variation due to adsorption of the sepiolite specimen was measured every 30 min with a formaldehyde testing instrument (PPM-400ST) at a controlled temperature of 296.15 K. The adsorption ratio, calculated using the equation 100(C 0 − C t)/C 0%, where C 0 is the initial formaldehyde concentration and C t is the formaldehyde concentration at time t, was used to evaluate the formaldehyde adsorption.
Results and discussion
Effect of Ca(OH)2 content
Based on the hardening mechanism of cement and concrete and because the sepiolitic raw material is rich in SiO2 and poor in CaO (Table 1), Ca(OH)2 should be introduced to form C-A-S-H or Al-tobermorite to enhance the strength of specimens (Sun et al., Reference Sun, Jing, Cai, Zhang, Zhao and Xu2015; Miao et al., Reference Miao, Jing, Pu and Zhang2018).
The flexural strengths of the specimens mixed with various Ca(OH)2 contents and cured at 473.15 K for 12 h are shown in Fig. 4. The flexural strength of the raw sepiolite (0 mass% Ca(OH)2) is rather low (2.4 MPa); with increasing additions of Ca(OH)2, the flexural strength of the specimens increases significantly. At 30 mass% Ca(OH)2, the flexural strength is maximized (11 MPa), which is almost four times greater than that of the sepiolite specimen at 0 mass% Ca(OH)2. This strength decreases with greater additions of Ca(OH)2.
Fig. 4. Effects of Ca(OH)2 content on the flexural strength of specimens cured at 473.15 K for 12 h.
The phase evolution within the matrix with increasing Ca(OH)2 was investigated by XRD (Fig. 5). The semi-quantitative mineralogical composition of the samples is listed in Table 2. The original sample consists mainly of sepiolite with minor quartz and talc, which gradually decrease with increasing Ca(OH)2 contents due to the reaction with Ca(OH)2. Concurrently, the C-A-S-H phases appear at 10 mass% Ca(OH)2 and Al-tobermorite begins to form after addition of 20 mass% Ca(OH)2 addition. Al-tobermorite may partly form from C-A-S-H phases because Al is contained in the raw material (Schreiner et al., Reference Schreiner, Goetz-Neunhoeffer, Neubauer and Jansen2020). Both C-A-S-H and Al-tobermorite are maximized when 30 mass% Ca(OH)2 is added. However, with further addition of Ca(OH)2, Al-tobermorite tends to disappear (Table 2). The excessive addition of Ca(OH)2 forms a protective coating on the surface of anhydrous particles and thus may inhibit further hydration. The formation of Al-tobermorite and C-A-S-H phases might be related to the observed increase in flexural strength, in agreement with previous work (Zhang et al., Reference Zhang, Jing and Fan2013). In addition, most sepiolite is still present in the matrix at 30 mass% Ca(OH)2 (the maximum strength) (see the inset in Fig. 5 and Table 2).
Fig. 5. XRD traces of samples containing various proportions of Ca(OH)2 and cured at 473.15 K for 12 h.
Table 2. Semi-quantitative composition of samples containing various Ca(OH)2 contents (%).
The FTIR spectra of the specimens with various Ca(OH)2 contents cured at 473.15 K for 12 h are shown in Fig. 6. With increasing Ca(OH)2 content, the bands corresponding to the stretching vibrations of sepiolite (1028, 799, 780 and 692 cm–1; Liang et al., Reference Liang, Xu, Sun, Wang, Sun, Sun and Qin2011) tend to weaken, and a new band appears at 970 cm–1, which might be attributed to Al-tobermorite and C-A-S-H phases (Miao et al., Reference Miao, Jing, Pu and Zhang2018). Therefore, in agreement with the XRD results, the addition of Ca(OH)2 initially contributes to the transformation of sepiolite and the formation of Al-tobermorite and C-A-S-H phases, which results in increased strength. With further addition of Ca(OH)2 (to 40 mass%), the band intensity at 970 cm–1 tends to decrease because excessive Ca(OH)2 addition limits hydration, which is in agreement with the XRD results (Fig. 5). The migration of the Si–O band from 1028 to 970 cm–1 might be related to the decrease in the degree of polymerization of [SiO4]4– units because the raw materials contain certain amounts of A13+ (Table 1), and thus the Si4+ in the [SiO4]4– tetrahedra may be partially replaced by Al3+ to form alumina polyhedrons (i.e. Al-tobermorite and C-A-S-H phases; Li et al., Reference Li, Zhang, Garbev, Beuchle and Monteiro2020). The band at 1028 cm–1 is still of high intensity, indicating that most sepiolite might survive after the hydrothermal reaction. Therefore, to synthesize the cemented sepiolite aggregate, the addition of 30 mass% Ca(OH)2 was used throughout this study.
Fig. 6. FTIR spectra of samples containing various proportions of Ca(OH)2 and cured at 473.15 K for 12 h.
Effect of curing conditions
The effects of curing time and temperature on the flexural strength were investigated. The strength evolution of specimens cured at 473.15 K for various curing times and cured for 6 h at various temperatures are shown in Fig. 7. The flexural strength increases rapidly up to 3 h curing time, reaching a maximum strength (>17 MPa) at 6 h, which is three times greater than that without curing (0 h). For greater curing times, the flexural strength declines gradually. The flexural strength is also influenced by the curing temperature. It increases gradually to >17 MPa until 473.15 K, and then decreases to 13 MPa at 513.15 K.
Fig. 7. Effects of (a) curing times and (b) temperatures on the flexural strength of specimens.
The XRD traces of specimens cured at 473.15 K for various curing times and specimens cured for 6 h at various temperatures are shown in Fig. 8 and the semi-quantitative composition is listed in Table 3. Sepiolite, quartz, talc and portlandite continuously decrease while some C-A-S-H phases and Al-tobermorite appears, suggesting more sepiolitic raw material and Ca(OH)2 transfer to C-A-S-H phases and Al-tobermorite for higher curing times or temperatures (Table 4). Comparison of these phase evolutions with the strength development shows that the C-A-S-H phases and Al-tobermorite formation enhance the strength. Al-tobermorite crystals will grow at longer curing times or higher curing temperatures, and the overgrowth of Al-tobermorite may loosen the matrix, thus exerting a minor influence on the strength. It should be noted that the sepiolite phase exists in all of the synthesized specimens (Table 4), while the Al-tobermorite formed at longer curing times and higher curing temperatures may lead to the formation of a tough matrix. As such, a cemented sepiolite aggregate material might be solidified hydrothermally with sepiolite clay powder.
Fig. 8. XRD patterns of solidified specimens at (a) various curing times and (b) various temperatures.
Table 3. Semi-quantitative composition of solidified specimens cured at 473.15 K for various curing times and cured for 6 h at various temperatures.
Table 4. Specific surface areas and pore volumes of specimens cured at 473.15 K for 0, 6 and 12 h.
The ESEM micrographs (Fig. 9) reveal the change in the phase morphology of the specimens cured at 473.15 K for 0 and 6 h. Before the hydrothermal treatment (0 h), a mass of fibrous sepiolite, rod-like talc and massive quartz can be observed clearly in the matrix. After 6 h of curing (Fig. 9b), some needle-like or tabular crystals of Al-tobermorite and flocculating C-A-S-H (Li et al., Reference Li, Zhang, Garbev, Beuchle and Monteiro2020) form, filling the spaces between the sepiolite fibres, thus enhancing the flexural strength and forming a cemented sepiolite aggregate. In addition, some mesopores form on the surface of the Al-tobermorite crystals (Fig. 9c) and intercrystallite mesopores appear between crosslinked Al-tobermorite crystals (Fig. 9d). Therefore, the formation of Al-tobermorite may not only enhance strength, but also increase mesoporosity.
Fig. 9. ESEM images of specimens cured at 473.15 K for (a) 0 h and (b–d) 6 h.
Adsorption performance and mechanism
The performance of the cemented sepiolite aggregate was tested for humidity-regulation and formaldehyde-adsorption properties in specimens with the addition of 30 mass% Ca(OH)2 and cured at 473.15 K for 0, 6 and 12 h.
Humidity-regulating capacity
Figure 10 displays the moisture adsorption/desorption performance of the specimens when the RH of the environment changed from 75% to 33% at 298.15 K for various curing times. At 0 h curing time, the specimen exhibits good moisture adsorption performance (moisture adsorption of 230 g m–2) due to the presence of the sepiolite in the matrix. After 6 h of curing time, maximum moisture adsorption occurred, exceeding 430 g m–2 – nearly twice that achieved without curing (0 h). This increase in moisture adsorption might be mainly due to the neoformed phases within the matrix, namely Al-tobermorite and C-A-S-H phases, and the mesopores formed on the surface and between the crystals of Al-tobermorite (Fig. 9). This suggests that the formed Al-tobermorite and C-A-S-H phases may not only enhance strength, but also improve the moisture adsorption/desorption (humidity-regulation) performance.
Fig. 10. Humidity-regulating properties of specimens cured at 473.15 K for various curing times.
A detailed investigation of the microstructure (porosity) evolution of the specimens with various curing times was conducted to study the improvement in humidity regulation. The pore-size distributions of the specimens cured for 0, 6 and 12 h are shown in Fig. 11 and their specific surface areas and pore volumes are listed in Table 4.
Fig. 11. Pore-size distributions of specimens cured for various curing times.
The specific surface areas of specimens without curing (0 h) is only 62.50 m2 g–1 and the total pore volume is 0.2168 cm3 g–1. After curing for 6 h, the specific surface area increases nearly twofold (111.48 m2 g–1) and the pore volume also increases markedly. For specimens cured for 12 h, the specific surface area and pore volume decrease to 101.781 m2 g–1 and 0.2800 cm3 g–1, respectively.
Before hydrothermal processing (0 h), the pore sizes range from 5 to 42 nm, corresponding to the spaces between sepiolite fibres (Fig. 11). After hydrothermal processing, the size of mesopores increases significantly. After curing for 6 h, more mesopores form, especially in the 4–9 nm range, which also results in the largest specific surface area (111.48 m2 g–1) and pore volume (0.3576 cm3 g–1; Table 5). A pore size of 4.2–7.1 nm might provide effective humidity regulation (Lan et al., Reference Lan, Jing and Li2017) and might explain why curing for 6 h yields the best moisture adsorption/desorption (humidity-regulation) performance. As was mentioned above, after 6 h of curing, abundant sepiolite crystals remained in the matrix, and at the same time Al-tobermorite formed. The Al-tobermorite might enhance strength and may also markedly improve pore-size distribution, specific surface area and pore volume, thus significantly promoting humidity regulation.
Table 5. Constants and coefficients of determination of the Langmuir model.
The water-vapour adsorption/desorption isotherms of the solidified sediment specimens cured for various times are shown in Fig. 12. The water-vapour adsorption/desorption without hydrothermal processing is low, with few changes between 0.3 and 0.8 relative water-vapour pressure (30–80% RH). After curing for 6 h, the water-vapour adsorption/desorption capacity was greatly improved and the amount of the adsorbed/desorbed water increased from ~20 to 146 mg g–1. The improvement in humidity regulation is probably due to the retained sepiolite and the formation of Al-tobermorite (Fig. 8 & Table 3), which in turn enhanced porosity (Figs 9 & 11). However, when the curing time extended to 12 h, the amount of adsorbed/desorbed water decreased to 117 mg g–1, probably because of the disappearance of sepiolite (Fig. 6) and the overgrowth of Al-tobermorite crystals, which might reduce mesopores.
Fig. 12. Water-vapour isotherms at 25°C for the specimens cured for various curing times. Solid and open symbols represent adsorption and desorption isotherms, respectively.
Formaldehyde-removal properties
Figure 13 shows the VOC (formaldehyde) adsorption ratio of specimens at various curing times. The greatest formaldehyde removal (70%) was observed in the sample without hydrothermal processing (0 h), followed by the sample cured for 6 h (68%), while the sample cured for 12 h adsorbed ~60% of the formaldehyde after 4 h. According to the peak intensity change with various curing times in the XRD traces (Fig. 8b), the formaldehyde removal seems to be related to the residual sepiolite content in the matrix (i.e. the more residual sepiolite, the greater the formaldehyde removal; Table 3). The formaldehyde removal seems not to depend on the mesopores because the specimens cured for 6 and 12 h have more mesopores than those without curing.
Fig. 13. Formaldehyde-adsorption capacity of specimens cured at 473.15 K for various curing times.
As was mentioned before, the sepiolite itself is also rich in micropores, and an investigation into the change in the pore-size distribution of the specimens at various curing times (0 and 12 h) was conducted (Fig. 14). The hydrothermal processing does not affect the micropore size, which varies between 0.35 and 0.75 nm in the uncured and the cured samples, and this might be mainly attributed to the microporous structure of sepiolite. Nevertheless, although after hydrothermal processing, the D(50) value of the specimens cured for 0 and 12 h is almost the same, there seem to be fewer micropores in the specimen cured for 12 h. This confirms the notion that the micropores probably originate from the sepiolite, not from the neoformed Al-tobermorite, which is also in accordance with the semi-quantitative XRD data (Table 3). With increasing curing time, the remaining sepiolite decreases due to its reaction to form Al-tobermorite (Fig. 8 & Table 3), which inevitably decreases the number of micropores within the matrix, thus resulting reduced formaldehyde removal.
Fig. 14. Pore-size distributions of specimens cured for various curing times.
As discussed above, it seems that the formaldehyde adsorption depends mainly on the inherent micropores of sepiolite, while the neoformed Al-tobermorite seems to be less important in this study. The experimental data regarding the humidity regulation and formaldehyde adsorption of the specimens cured at 473.15 K for 6 h were used to fit the Langmuir isotherm models according to Eq. (1):
$$\log \left( {q_c-q_e} \right) = {\rm log}q_e-\displaystyle{{\rm k} \over {2.303}}t$$where qc and qe (mg g–1) are the uptakes of moisture or formaldehyde per unit mass of adsorbent at time t and saturation time and k is the Langmuir constant related to the rate of adsorption. The calculated constants are presented in Table 5.
The calculated Langmuir isotherm gives a good fit only for formaldehyde adsorption, with a goodness of fit value of R 2 >0.99 (Table 5). By contrast, the coefficient of determination for moisture adsorption according to the Langmuir model is low (R 2 = 0.88). This suggests that the formaldehyde adsorption isotherm belongs to type IV, and thus its adsorption depends mainly on the micropores in the matrix, which is in agreement with the results of this and previous work (Chen et al., Reference Chen and Yang1996). According to our previous work, a pore size of 4.2–7.1 nm in the materials led to effective moisture adsorption/desorption performance (Lan et al., Reference Lan, Jing and Li2017), which might explain why the moisture adsorption data the had a poor fit with type I adsorption isotherm.
Conclusions
To extend the applications of sepiolite, hydrothermal synthesis of a cemented sepiolite aggregate for humidity regulation and formaldehyde adsorption was conducted, and the results are summarized as follows:
(1) The sepiolite was solidified into a cemented aggregate with maximum flexural strength of >17 MPa, in which most sepiolite still remained in the matrix due to the low-temperature hydrothermal synthesizing approach (≤473.15 K). The formation of C-A-S-H phases and Al-tobermorite enhanced the strength of the specimens, and Ca(OH)2 addition, curing temperature and time influenced their formation.
(2) The porosity (pore-size distribution, specific surface area and pore volume) of the specimens is the key factor for humidity regulation. The mesopores from to the sepiolite and the C-A-S-H and Al-tobermorite phases formed increased the moisture adsorption/desorption performance significantly. The moisture adsorption of the specimen cured for 6 h exceeded 430 g m–2, which is nearly twice that of the original sepiolite specimen without hydrothermal processing.
(3) Similarly to the original sepiolite, the produced material also displayed good formaldehyde removal (60–68%). However, by contrast to the humidity regulation, the micropores originating mainly from the remaining sepiolite in the matrix and not from the neoformed C-A-S-H phases and Al-tobermorite controlled formaldehyde removal.
Sepiolite can be converted hydrothermally into a cemented sepiolite aggregate with the capacity for humidity regulation and formaldehyde removal, and this aggregate might be used as a building material to provide comfortable and safe living conditions indoors.
Financial support
This work was supported by the National Natural Science Foundation of China (nos. 51872206, 51272180) and the National Key R&D Program of China (2017YFC 0507004).
